“Self-es­teem” is distortionary

Algon23 Nov 2025 23:59 UTC
15 points
3 comments2 min readLW link

Cy­ber­bud­dhist Jar­gon 1.0

lsusr23 Nov 2025 23:39 UTC
50 points
21 comments7 min readLW link

Find­ing the un­cer­tainty vec­tor in GPT2-scale transformers

larry-dial23 Nov 2025 23:34 UTC
9 points
0 comments10 min readLW link

Stop Ap­ply­ing And Get To Work

23 Nov 2025 22:50 UTC
221 points
58 comments2 min readLW link

Halfhaven Digest #5

Taylor G. Lunt23 Nov 2025 21:57 UTC
15 points
0 comments3 min readLW link

Emo­tions, Fabricated

Dentosal23 Nov 2025 21:57 UTC
4 points
0 comments2 min readLW link

I’ll be sad to lose the puzzles

Ruby23 Nov 2025 19:37 UTC
112 points
21 comments2 min readLW link

Show Re­view: Masquerade

johnswentworth23 Nov 2025 19:20 UTC
41 points
2 comments3 min readLW link

AI Sen­tience and Welfare Misal­ign­ment Risk

edgecase6423 Nov 2025 18:22 UTC
14 points
3 comments8 min readLW link

If you can­not be good, at least be bad correctly

beyarkay (Boyd Kane)23 Nov 2025 17:51 UTC
17 points
1 comment2 min readLW link
(boydkane.com)

Please Mea­sure Ver­ifi­ca­tion Burden

Quinn23 Nov 2025 17:25 UTC
17 points
4 comments4 min readLW link

Sols­tice Sin­ga­long Watch Party

23 Nov 2025 16:36 UTC
11 points
0 comments1 min readLW link

Busk­ing Practice

jefftk23 Nov 2025 15:20 UTC
16 points
0 comments1 min readLW link
(www.jefftk.com)

The Enemy Gets The Last Hit

J Bostock23 Nov 2025 12:22 UTC
47 points
5 comments3 min readLW link

A list of peo­ple who could’ve started a nu­clear war, but chose not to

Mikhail Samin23 Nov 2025 9:25 UTC
28 points
5 comments5 min readLW link

Tra­di­tional Food

lsusr23 Nov 2025 8:07 UTC
109 points
10 comments9 min readLW link

Me­mories of a Bri­tish Board­ing School #2.5

Ben Pace23 Nov 2025 7:54 UTC
23 points
2 comments2 min readLW link

Dipole Nature

alkjash23 Nov 2025 7:24 UTC
40 points
2 comments5 min readLW link
(radimentary.wordpress.com)

What kind of per­son is Deep­Seek’s founder, Liang Wen­feng? An an­swer from his old uni­ver­sity class­mate.

L.M.Sherlock23 Nov 2025 4:54 UTC
92 points
0 comments4 min readLW link
(lmsherlock.substack.com)

Com­ment on Nat­u­ral Emer­gent Misal­ign­ment Paper by Anthropic

Simon Lermen23 Nov 2025 4:21 UTC
21 points
0 comments4 min readLW link

How to throw parties

RobertM23 Nov 2025 3:59 UTC
22 points
0 comments5 min readLW link

Stream of Con­scious­ness as a Scaf­fold­ing Skill

Screwtape23 Nov 2025 3:31 UTC
33 points
2 comments4 min readLW link

Liter­acy is De­creas­ing Among the In­tel­lec­tual Class

Taylor G. Lunt23 Nov 2025 3:08 UTC
37 points
29 comments10 min readLW link

Mar­ket Logic II

abramdemski23 Nov 2025 1:41 UTC
24 points
3 comments7 min readLW link

You can just do things: 5 frames

Algon23 Nov 2025 0:43 UTC
54 points
3 comments3 min readLW link

Easy vs Hard Emo­tional Vulnerability

johnswentworth23 Nov 2025 0:15 UTC
34 points
25 comments2 min readLW link

Why your sports car isn’t a race­car (trade­offs ev­ery­where)

Ruby22 Nov 2025 23:23 UTC
29 points
0 comments5 min readLW link

As­sorted Thoughts on “Pivot­ing” to AI

Trevor Hill-Hand22 Nov 2025 21:17 UTC
12 points
1 comment4 min readLW link

OpenAI Locks Down San Fran­cisco Offices Fol­low­ing Alleged Threat From Activist

Matrice Jacobine22 Nov 2025 19:33 UTC
40 points
0 comments4 min readLW link
(www.wired.com)

Sorry, I still think kid­ney dona­tion makes no sense for an EA

nicholashalden22 Nov 2025 18:10 UTC
6 points
4 comments1 min readLW link
(substack.com)

Au­to­matic alt text generation

TurnTrout22 Nov 2025 17:57 UTC
27 points
1 comment1 min readLW link
(turntrout.com)

My frus­tra­tions: AI doom

Dentosal22 Nov 2025 14:59 UTC
2 points
0 comments2 min readLW link

In­tro­spec­tion in LLMs: A Pro­posal For How To Think About It, And Test For It

Christopher Ackerman22 Nov 2025 14:52 UTC
23 points
4 comments7 min readLW link

AI Red Lines: A Re­search Agenda

Charbel-Raphaël22 Nov 2025 8:41 UTC
30 points
1 comment5 min readLW link

Book Re­view: Wizard’s Hall

Screwtape22 Nov 2025 7:38 UTC
96 points
4 comments5 min readLW link

Be Naughty

habryka22 Nov 2025 6:35 UTC
99 points
11 comments4 min readLW link

Mar­ket Logic I

abramdemski22 Nov 2025 6:01 UTC
36 points
2 comments5 min readLW link

The AI 2027 Re­port Is Not Backed Up by Evidence

Oscar Davies22 Nov 2025 5:23 UTC
−17 points
9 comments4 min readLW link

LLM Sys­tems for Liter­a­ture-Based Scien­tific Discovery

Carly Turini22 Nov 2025 4:48 UTC
1 point
0 comments1 min readLW link

An­i­mal welfare con­cerns are dom­i­nated by post-ASI futures

RobertM22 Nov 2025 4:08 UTC
28 points
1 comment4 min readLW link

Ha­bit­ual men­tal mo­tions might ex­plain why peo­ple are con­tent to get old and die

Ruby22 Nov 2025 2:52 UTC
19 points
1 comment7 min readLW link

D&D.Sci Thanks­giv­ing: the Fes­ti­val Feast

aphyer22 Nov 2025 2:26 UTC
41 points
15 comments2 min readLW link

Di­plo­macy dur­ing AI takeoff

Nikola Jurkovic22 Nov 2025 2:12 UTC
18 points
3 comments2 min readLW link
(nikolajurkovic.substack.com)

Ab­stract ad­vice to re­searchers tack­ling the difficult core prob­lems of AGI alignment

TsviBT22 Nov 2025 0:53 UTC
130 points
10 comments8 min readLW link

Easy Op­por­tu­nity to Help Many Animals

Bentham's Bulldog21 Nov 2025 23:03 UTC
10 points
0 comments1 min readLW link

Why Not Just Train For In­ter­pretabil­ity?

johnswentworth21 Nov 2025 22:08 UTC
56 points
12 comments4 min readLW link

Com­plain­ing about my in­abil­ity to fo­cus on un­in­ter­est­ing things

Dentosal21 Nov 2025 20:34 UTC
5 points
3 comments2 min readLW link

Models not mak­ing it clear when they’re role­play­ing seems like a fairly big issue

williawa21 Nov 2025 20:23 UTC
16 points
3 comments6 min readLW link

Nat­u­ral Emer­gent Misal­ign­ment from Re­ward Hacking

Algon21 Nov 2025 20:20 UTC
12 points
0 comments3 min readLW link
(www.anthropic.com)

Nat­u­ral emer­gent mis­al­ign­ment from re­ward hack­ing in pro­duc­tion RL

21 Nov 2025 20:00 UTC
258 points
32 comments9 min readLW link